Comprehensive evaluation index determination method and device, storage medium and electronic equipment

Abstract

The embodiment of the invention provides a comprehensive evaluation index determination method and device, a storage medium and electronic equipment, and the method comprises the steps: sampling fin parameters of a cold plate fin to obtain s groups of first sample points, the fin parameters comprise the fin length, the fin width and the fin spacing, and the s groups of first sample points comprise the fin length, the fin width and the fin spacing; the s groups of first sample points are used for indicating numerical value combinations of various fin parameters, and s is a positive integer; performing a fitting test on the cold plate fin according to the s groups of first sample points, and determining a harmonic parameter of a comprehensive evaluation index of the cold plate fin according to an obtained fitting test result, the comprehensive evaluation index being used for performing comprehensive evaluation on thermal performance and hydraulic performance of the cold plate fin, the harmonic parameters are used for indicating the importance degree of the dimensionless temperature gradient and the flow resistance loss of the cold plate fin, the dimensionless temperature gradient is used for evaluating the thermal performance of the cold plate fin, and the flow resistance loss is used for evaluating the hydraulic performance of the cold plate fin; and determining comprehensive evaluation indexes of the cold plate fins according to the harmonic parameters.

Description

Technical Field

[0001] The embodiments of the present application relate to the field of computers, and more specifically, to a method and device for determining a comprehensive evaluation index, a storage medium, and an electronic device. Background Art

[0002] At present, many designs of liquid-cooled cold plates have begun to introduce multi-objective optimization design methods for optimal design. Among the more common optimization schemes, dual-objective optimization designs are usually adopted simultaneously in the two directions of minimizing flow resistance and maximizing heat transfer performance. Sample points are obtained by local sampling combined with experimental verification or numerical calculation, and then the optimal point set of dual-objective optimization is obtained through fitting methods such as neural networks, genetic algorithms or response surface optimization, and then manual selection is performed in the optimal point set.

[0003] The current mainstream multi-objective optimization algorithms usually use a fixed algorithm structure to design sampling points generated by manual enumeration. The sample capacity is not comprehensive and is prone to local omissions or over-concentration. When the algorithm is used for data fitting, overfitting will occur, thereby reducing the error between the multi-objective optimization and the actual state. In addition, how to make full use of the point set generated by multi-objective optimization and select the combination of the best state points from it has exceeded the scope of the algorithm solution for multi-objective optimization. The current selection method is mostly manual selection, and the quantitative basis is relatively poor.

[0004] Regarding the related technologies, in the design optimization process for liquid-cooled cold plates, a dual-objective optimization design is performed simultaneously in two directions, namely, minimizing flow resistance and maximizing heat exchange performance. However, it is difficult to determine the technical problem of the optimal state point, and no effective solution has been proposed yet. Summary of the invention

[0005] The embodiments of the present application provide a method and device for determining a comprehensive evaluation index, a storage medium, and an electronic device, so as to at least solve the problem in the related art that, in the design optimization process of a liquid-cooled cold plate, dual-objective optimization design is performed in two directions simultaneously for minimizing flow resistance and maximizing heat exchange performance, and it is difficult to determine the optimal state point.

[0006] According to an embodiment of the present application, a method for determining a comprehensive evaluation index is provided, comprising: sampling fin parameters of a cold plate fin to obtain s groups of first sample points, wherein the fin parameters include: fin length, fin width and fin spacing, and the s groups of first sample points are all used to indicate a numerical combination of multiple fin parameters, and s is a positive integer; performing a fitting test on the cold plate fin according to the s groups of first sample points, and determining a harmonic parameter of the comprehensive evaluation index of the cold plate fin according to the obtained fitting test result, wherein the comprehensive evaluation index is used to comprehensively evaluate the thermal performance and hydraulic performance of the cold plate fin, and the harmonic parameter is used to indicate the importance of the dimensionless temperature gradient and flow resistance loss of the cold plate fin, the dimensionless temperature gradient is used to evaluate the thermal performance of the cold plate fin, and the flow resistance loss is used to evaluate the hydraulic performance of the cold plate fin; the comprehensive evaluation index of the cold plate fin is determined according to the harmonic parameter.

[0007] In an exemplary embodiment, the fin parameters of the cold plate fins are sampled to obtain s groups of first sample points, including: determining the value ranges of the multiple fin parameters respectively, and stratifying the multiple value ranges into s layers, wherein the s sub-intervals obtained after stratification of each value range are of equal length, and the multiple fin parameters correspond one-to-one to the multiple value ranges; performing random sampling in the s sub-intervals of each fin parameter to obtain s random sampling points corresponding to each fin parameter; a combination step: selecting a first random sampling point from the s random sampling points corresponding to each fin parameter by a random number seed, and combining the multiple first random sampling points to obtain a group of sample points, wherein the multiple fin parameters correspond one-to-one to the multiple first random sampling points; looping the combination step s times to obtain s groups of second sample points, wherein the s second random sampling points corresponding to the same fin parameter in the s groups of second sample points are different from each other; determining the s groups of first sample points according to the s groups of second sample points.

[0008] In an exemplary embodiment, determining the s groups of first sample points based on the s groups of second sample points includes: taking multiple of the first random sampling points of the second sample points as spatial coordinates, respectively calculating the spatial distances between any two groups of second sample points, and obtaining multiple first spatial distances; determining the size of a second spatial distance among the multiple first spatial distances and a preset minimum distance, wherein the second spatial distance is the smallest among the multiple first spatial distances; and determining the s groups of second sample points as the s groups of first sample points when the second spatial distance is greater than the preset minimum distance.

[0009] In an exemplary embodiment, after determining the size of the second spatial distance among the multiple first spatial distances and the preset minimum distance, the method also includes: a first adjustment step: when the second spatial distance is less than the preset minimum distance, adjusting multiple of the first random sampling points of the two groups of second sample points corresponding to the second spatial distance so that the spatial distance between the two groups of second sample points corresponding to the second spatial distance is greater than the preset minimum distance; an updating step: recalculating the spatial distance between any two groups of second sample points in the s groups of second sample points, and updating the second spatial distance according to the calculation result; and looping the first adjustment step and the updating step until the second spatial distance is greater than the preset minimum distance.

[0010] In an exemplary embodiment, before determining the reconciliation parameter of the comprehensive evaluation index of the cold plate fin according to the obtained fitting test result, the method further includes: determining the target expression of the comprehensive evaluation index as: Among them, ES is the comprehensive evaluation index, Nu is the dimensionless temperature gradient, Δp is the flow resistance loss, Nua is the reference value of the dimensionless temperature gradient, Δpa is the reference value of the flow resistance loss, and α and β are both the harmonic parameters.

[0011] In an exemplary embodiment, the fitting test is performed on the cold plate fin according to the first sample point of the s group, and the harmonic parameter of the comprehensive evaluation index of the cold plate fin is determined according to the obtained fitting test result, including: performing a fitting test on the cold plate fin according to the first sample point of the s group, determining the heat exchange target temperature T of the cold plate fin i and cost parameter S i , wherein the heat exchange target temperature is used to indicate the heat exchange performance of the cold plate fin, the thermal performance includes the heat exchange performance, and the cost parameter is used to indicate the construction cost of the cold plate fin, T i is the heat exchange target temperature of the cold plate fin corresponding to the first sample point of the i-th group, S i is the cost parameter of the cold plate fin corresponding to the first sample point of the i-th group; determining the engineering priority sequence number C of the first sample point of the s-th group according to the engineering requirement information of the cold plate fin, the heat exchange target temperature and the cost parameter TSi , where C TSi is the engineering priority sequence number of the first sample point of the i-th group, i is a positive integer; the initial value α0 is set for the harmonic parameter α, and the initial value β0 is set for the harmonic parameter β to obtain the first harmonic parameter; a fitting test is performed based on the first sample point of the s group to calculate the dimensionless temperature gradient Nu of the cold plate fin i and flow resistance loss Δp i, wherein the fitting test results include: the heat exchange target temperature T i , the cost parameter S i , the dimensionless temperature gradient Nu i and the flow resistance loss Δp i , Nu i is the dimensionless temperature gradient of the cold plate fin corresponding to the first sample point of the i-th group, Δp i is the flow resistance loss of the cold plate fin corresponding to the first sample point of the i-th group; according to the first harmonic parameter, the target expression, the dimensionless temperature gradient Nu i and the flow resistance loss Δp i Calculate the first index value of the comprehensive evaluation index, and determine the index priority sequence number C of the s groups of first sample points according to the s first index values ESi , where C ESi is the index priority sequence number of the first sample point in the i-th group; according to C TSi With C ESi The first harmonic parameter is adjusted according to the absolute difference of , so as to obtain the harmonic parameter.

[0012] In an exemplary embodiment, the TSi With C ESi The method comprises: a second adjustment step: when the maximum value among the plurality of absolute differences is greater than the difference judgment standard, adjusting the first harmonic parameter according to the maximum value to obtain a second harmonic parameter; a calculation step: calculating the second harmonic parameter, the target expression, the dimensionless temperature gradient Nu according to the second harmonic parameter, the target expression, the dimensionless temperature gradient Nu, and the dimensionless temperature gradient Nu. i and the flow resistance loss Δp i Calculate the second index value of the comprehensive evaluation index, and re-determine the index priority sequence number C of the s groups of first sample points according to the s second index values ESi ; Determination step: recalculate multiple absolute differences, and redetermine the maximum value among the multiple absolute differences, and determine the size of the maximum value and the difference judgment standard; loop the second adjustment step, the calculation step and the determination step until the maximum value is less than or equal to the difference judgment standard, and determine the final second harmonic parameter as the harmonic parameter.

[0013] According to another embodiment of the present application, a device for determining a comprehensive evaluation index is provided, comprising: a sampling module, used to sample fin parameters of a cold plate fin to obtain s groups of first sample points, wherein the fin parameters include: fin length, fin width and fin spacing, and the s groups of first sample points are all used to indicate a numerical combination of multiple fin parameters, and s is a positive integer; a testing module, used to perform a fitting test on the cold plate fin according to the s groups of first sample points, and determine a harmonic parameter of the comprehensive evaluation index of the cold plate fin according to the obtained fitting test result, wherein the comprehensive evaluation index is used to comprehensively evaluate the thermal performance and hydraulic performance of the cold plate fin, and the harmonic parameter is used to indicate the importance of the dimensionless temperature gradient and flow resistance loss of the cold plate fin, the dimensionless temperature gradient is used to evaluate the thermal performance of the cold plate fin, and the flow resistance loss is used to evaluate the hydraulic performance of the cold plate fin; a determination module, used to determine the comprehensive evaluation index of the cold plate fin according to the harmonic parameter.

[0014] According to another embodiment of the present application, a computer-readable storage medium is provided, in which a computer program is stored, wherein the computer program is configured to execute the steps of any of the above method embodiments when run.

[0015] According to another embodiment of the present application, an electronic device is provided, including a memory and a processor, wherein the memory stores a computer program, and the processor is configured to run the computer program to execute the steps in any one of the above method embodiments.

[0016] According to another embodiment of the present application, a computer program product is provided, including a computer program, and when the computer program is executed by a processor, the steps of the method described in each embodiment of the present application are implemented.

[0017] Through the present application, a method for determining a comprehensive evaluation index is proposed. First, the fin parameters of the cold plate fin are sampled to obtain s groups of first sample points. The fin parameters include: fin length, fin width and fin spacing. The s groups of first sample points are used to indicate different combinations of fin parameters, and s is a positive integer; a fitting test is performed on the cold plate fin according to the s groups of first sample points, and a harmonic parameter of a comprehensive evaluation index for comprehensively evaluating the thermal performance and hydraulic performance of the cold plate fin is determined according to the fitting test results. The harmonic parameter is used to indicate the importance of the dimensionless temperature gradient and flow resistance loss of the cold plate fin. The dimensionless temperature gradient The degree is used to evaluate the thermal performance, and the flow resistance loss is used to evaluate the hydraulic performance; finally, the comprehensive evaluation index of the cold plate fin is determined according to the harmonized parameter; using the above scheme, a dual-objective dimensionality reduction method is proposed, and a new comprehensive evaluation index is created according to the thermal performance index and the hydraulic performance index, so that the cold plate can be optimized accurately and quickly with a single objective through the comprehensive evaluation index to determine the optimal state point; thereby solving the problem in the related technology that the design optimization process of the liquid-cooled cold plate is difficult to determine the optimal state point by simultaneously performing dual-objective optimization design in the two directions of minimizing the flow resistance and maximizing the heat exchange performance. BRIEF DESCRIPTION OF THE DRAWINGS

[0018] Figure 1 It is a hardware structure block diagram of a computer terminal according to a method for determining a comprehensive evaluation index in an embodiment of the present application;

[0019] Figure 2 is a flow chart of a method for determining a comprehensive evaluation index according to an embodiment of the present application;

[0020] Figure 3 is a flow chart of an optional improved Latin hypercube sampling method according to an embodiment of the present application;

[0021] Figure 4 is a schematic diagram of a workflow of an optional priority sequence method according to an embodiment of the present application;

[0022] Figure 5 is a schematic diagram of an optional cold plate shape according to an embodiment of the present application;

[0023] Figure 6 It is a structural block diagram of a method for determining a comprehensive evaluation index according to an embodiment of the present application. DETAILED DESCRIPTION

[0024] The embodiments of the present application will be described in detail below with reference to the accompanying drawings and in combination with the embodiments.

[0025] It should be noted that the terms "first", "second", etc. in the specification and claims of this application and the above-mentioned drawings are used to distinguish similar objects, and are not necessarily used to describe a specific order or sequence.

[0026] The method embodiments provided in the embodiments of the present application can be executed in a computer terminal or a similar computing device. Taking running on a computer terminal as an example, Figure 1 1 is a hardware structure block diagram of a computer terminal for determining a comprehensive evaluation index in an embodiment of the present application. Figure 1 As shown, the computer terminal may include one or more ( Figure 1 Only one is shown in the figure) a processor 102 (the processor 102 may include but is not limited to a processing device such as a microprocessor MCU or a programmable logic device FPGA) and a memory 104 for storing data, wherein the above-mentioned computer terminal may also include a transmission device 106 and an input / output device 108 for communication functions. It can be understood by those skilled in the art that Figure 1 The structure shown is only for illustration and does not limit the structure of the above-mentioned computer terminal. Figure 1 More or fewer components as shown, or with Figure 1 Different configurations are shown.

[0027] The memory 104 can be used to store computer programs, for example, software programs and modules of application software, such as the computer program corresponding to the method for determining the comprehensive evaluation index in the embodiment of the present application. The processor 102 executes various functional applications and data processing by running the computer program stored in the memory 104, that is, to implement the above method. The memory 104 may include a high-speed random access memory, and may also include a non-volatile memory, such as one or more magnetic storage devices, flash memory, or other non-volatile solid-state memory. In some examples, the memory 104 may further include a memory remotely arranged relative to the processor 102, and these remote memories may be connected to the computer terminal via a network. Examples of the above-mentioned network include, but are not limited to, the Internet, an intranet, a local area network, a mobile communication network, and combinations thereof.

[0028] The transmission device 106 is used to receive or send data via a network. The specific example of the above network may include a wireless network provided by a communication provider of a computer terminal. In one example, the transmission device 106 includes a network adapter (Network Interface Controller, referred to as NIC), which can be connected to other network devices through a base station so as to communicate with the Internet. In one example, the transmission device 106 can be a radio frequency (RF) module, which is used to communicate with the Internet wirelessly.

[0029] In related technologies, with the surge in demand for chip computing power, liquid cooling has become the future development trend of high-power cooling technology, and cold plate liquid cooling is still the mainstream technical direction of liquid cooling development due to its reliability and easy maintenance. Faced with components such as CPUs and GPUs with ultra-high power consumption of more than 500W, most cold plate design solutions adopt ways to enhance the thermal hydraulic performance of the cold plate, such as increasing the fin density of the cold plate and increasing the flow rate of the coolant fluid. However, at the same time, the flow resistance of the coolant flowing through the cold plate increases, and the pump power increases accordingly, so the heat exchange efficiency is reduced.

[0030] In many applications, the improvement of the cooling and heat transfer performance of the cold plate is accompanied by an increase in flow resistance. Therefore, it is necessary to determine an optimal balance point of flow and heat transfer for different application scenarios, thereby determining the parameters and specific form of the cold plate design. As the configuration and flow state of the cold plate become increasingly complex, the use of manual enumeration methods for testing or simulation requires a lot of time and computing resources, and the selected optimization point may not be the actual optimal point because the enumeration samples cannot fully cover the entire sampling space.

[0031] Different from academic problems, when determining the optimal parameter of engineering products, in addition to considering the theoretical optimal balance of thermal and hydraulic performance, practical engineering issues such as product compatibility, process complexity and product cost also need to be considered. These factors also need to be considered within the scope of practical parameter priority. However, the current optimization algorithms are all aimed at solving completely theoretical problems and fail to quantitatively evaluate engineering factors. Moreover, the final result of the existing multi-objective optimization is a series of point sets. There is currently no systematic method for further optimization based on this.

[0032] For example, a GPU card has a high power consumption GPU, so the cold plate performance design requirements are relatively high. The cold plate shape is as follows Figure 5 As shown in the figure, due to the limitation of the overall liquid cooling efficiency, the pump power of the coolant flowing through the node is also required to be minimized. Therefore, in the process of parametric design of the cold plate, it is necessary to find a parameter combination with an optimal balance point, at which the contradiction between the flow resistance requirement and the heat transfer requirement can be balanced, and the cold plate can ensure both high heat transfer performance and low flow resistance level.

[0033] In order to achieve the above-mentioned design state, a method for determining a comprehensive evaluation index is provided in this embodiment. Figure 2 is a flow chart of a method for determining a comprehensive evaluation index according to an embodiment of the present application, such as Figure 2 As shown, the method includes:

[0034] Step S202, sampling the fin parameters of the cold plate fins to obtain s groups of first sample points, wherein the fin parameters include: fin length, fin width and fin spacing, and the s groups of first sample points are used to indicate a numerical combination of multiple fin parameters, and s is a positive integer;

[0035] Step S204, performing a fitting test on the cold plate fin according to the s group of first sample points, and determining a harmonic parameter of a comprehensive evaluation index of the cold plate fin according to the obtained fitting test result, wherein the comprehensive evaluation index is used to comprehensively evaluate the thermal performance and hydraulic performance of the cold plate fin, and the harmonic parameter is used to indicate the importance of the dimensionless temperature gradient and flow resistance loss of the cold plate fin, the dimensionless temperature gradient is used to evaluate the thermal performance of the cold plate fin, and the flow resistance loss is used to evaluate the hydraulic performance of the cold plate fin;

[0036] Step S206, determining the comprehensive evaluation index of the cold plate fin according to the blending parameter.

[0037] Through the above scheme, a method for determining a comprehensive evaluation index is proposed. First, the fin parameters of the cold plate fin are sampled to obtain s groups of first sample points. The fin parameters include: fin length, fin width and fin spacing. The s groups of first sample points are used to indicate different combinations of fin parameters, and s is a positive integer; a fitting test is performed on the cold plate fin according to the s groups of first sample points, and a harmonic parameter of a comprehensive evaluation index for comprehensively evaluating the thermal performance and hydraulic performance of the cold plate fin is determined according to the fitting test results. The harmonic parameter is used to indicate the importance of the dimensionless temperature gradient and flow resistance loss of the cold plate fin. The dimensionless temperature The gradient is used to evaluate the thermal performance, and the flow resistance loss is used to evaluate the hydraulic performance; finally, the comprehensive evaluation index of the cold plate fin is determined according to the harmonized parameter; using the above scheme, a dual-objective dimensionality reduction method is proposed, and a new comprehensive evaluation index is created according to the thermal performance index and the hydraulic performance index, so that the cold plate can be optimized accurately and quickly with a single objective through the comprehensive evaluation index to determine the optimal state point; thereby solving the problem in the related technology that the design optimization process for the liquid-cooled cold plate is difficult to determine the optimal state point by simultaneously performing dual-objective optimization design in the two directions of minimizing the flow resistance and maximizing the heat exchange performance.

[0038] In an exemplary embodiment, the fin parameters of the cold plate fins are sampled to obtain s groups of first sample points, including: determining the value ranges of the multiple fin parameters respectively, and stratifying the multiple value ranges into s layers, wherein the s sub-intervals obtained after stratification of each value range are of equal length, and the multiple fin parameters correspond one-to-one to the multiple value ranges; performing random sampling in the s sub-intervals of each fin parameter to obtain s random sampling points corresponding to each fin parameter; a combination step: selecting a first random sampling point from the s random sampling points corresponding to each fin parameter by a random number seed, and combining the multiple first random sampling points to obtain a group of sample points, wherein the multiple fin parameters correspond one-to-one to the multiple first random sampling points; looping the combination step s times to obtain s groups of second sample points, wherein the s second random sampling points corresponding to the same fin parameter in the s groups of second sample points are different from each other; determining the s groups of first sample points according to the s groups of second sample points.

[0039] Before using machine learning methods for multi-objective optimization design, it is necessary to determine the sampling points used for training the optimization algorithm, and the selection of sample point input parameters needs to ensure complete coverage of the entire sampling space to ensure the accuracy of the machine learning algorithm fitting the actual state prediction representation. The sample points in the sample space usually obey the distribution of a certain probability model, such as normal distribution, Poisson distribution, etc. Direct sampling in the sample space is prone to omission of marginal intervals, which in turn affects the effect and quality of subsequent algorithm training data.

[0040] Therefore, the embodiment of the present application adopts the Latin hypercube sampling method based on multi-level stratified sampling, and improves the mode of minimizing the distance of the sampling results in the multidimensional vector space. The specific operation method is as follows:

[0041] 1. Select the key parameters of the cold plate fin design (i.e. the above-mentioned fin parameters): fin length L, fin width W and fin spacing D, and determine the value range of L, W and D based on the space limitation of the cold plate design, process limitation and preliminary evaluation of thermal hydraulic performance: [L min , L max ]、[W min , W max ]、[D min , D max ] (i.e., multiple value ranges);

[0042] 2. The three fin parameters are layered in their respective intervals. The number of layers is set to s. Then the intervals of the three key parameters can be divided into s sub-intervals, which are:

[0043] [L min1 , Lmax1 ],[L min2 , L max2 ],…,[L mins , L maxs ];

[0044] [W min1 , W max1 ],[W min2 , W max2 ],…,[W mins , W maxs ];

[0045] [D min1 , D max1 ],[D min2 , D max2 ], ..., [D mink , D maxk ];

[0046] 3. Perform random sampling in each level interval of each parameter, obtain s random sample points for each of the three parameters, generate random number seeds using the Monte Carlo algorithm, and use the random number seeds to randomly combine the sample points of L, W, and D, thereby generating a combination of s groups of random sample points (L i , W i , D i ) (i.e., the second sample point of the above-mentioned group s).

[0047] For example, for the fin parameter L, firstly, random sampling is performed in the s level intervals of L to obtain s random sample points, and then a random number seed such as 0.6 is generated by using the Monte Carlo algorithm. Then, the value range of L is taken as a whole, and a target sub-interval is determined in the value range according to the random number seed. Then, the random sample point located in the target sub-interval among the s random sample points is determined as the first random sample point. Then, for the fin parameters W and D, a first random sample point is selected in the same way, and these three first random sample points are combined into a group of sample points.

[0048] Through this embodiment, random sampling is performed to completely cover the entire sampling space, thereby avoiding omission of edge intervals, thereby ensuring the effect and quality of subsequent algorithm training data.

[0049] Based on the above steps, determining the s groups of first sample points according to the s groups of second sample points includes: taking multiple of the first random sampling points of the second sample points as spatial coordinates, respectively calculating the spatial distances between any two groups of second sample points, and obtaining multiple first spatial distances; determining the size of a second spatial distance among the multiple first spatial distances and a preset minimum distance, wherein the second spatial distance is the smallest among the multiple first spatial distances; and determining the s groups of second sample points as the s groups of first sample points when the second spatial distance is greater than the preset minimum distance.

[0050] After obtaining s groups of second sample points, it is necessary to calculate the space distance L between each group of sample points in the sample space. ij Calculate, that is Thus, multiple first spatial distances are obtained. The values ​​of the first spatial distances are compared and sorted, and the minimum value L of the sample point distance is determined accordingly. min (i.e. the second spatial distance mentioned above). s (i.e. the above-mentioned preset minimum distance) for comparison, if L min >L s , it means that the sampling result meets the minimum distance requirement and can be used for subsequent model training.

[0051] Optionally, after determining the size of the second spatial distance among the multiple first spatial distances and the preset minimum distance, the method also includes: a first adjustment step: when the second spatial distance is smaller than the preset minimum distance, adjusting multiple of the first random sampling points of the two groups of second sample points corresponding to the second spatial distance so that the spatial distance between the two groups of second sample points corresponding to the second spatial distance is greater than the preset minimum distance; an updating step: recalculating the spatial distance between any two groups of second sample points among the s groups of second sample points, and updating the second spatial distance according to the calculation result; and looping the first adjustment step and the updating step until the second spatial distance is greater than the preset minimum distance.

[0052] On the other hand, if L min <L s , it means that the sampling result does not meet the minimum distance requirement, and it is necessary to adjust the position between points i and j in the sampling space. After the adjustment, the minimum distance comparison of all distance values ​​is performed until all sample space points meet the minimum distance requirement.

[0053] Specifically, the positions of the two points can be adjusted in the following manner: the coordinate points of the two sampling points that do not meet the requirements are adjusted in the opposite direction to increase the spatial distance between the two sampling points, and the adjustment is stopped after the spatial distance exceeds the preset minimum distance, and the sampling point information is updated according to the adjusted spatial position.

[0054] In an optional embodiment, the present application proposes an improved Latin hypercube sampling method, the process of which is as follows: Figure 3 As shown, the following steps are included:

[0055] Step 3.1, start;

[0056] Step 3.2: Determine the range of L, W, and D [L min , L max ]、[W min , W max ]、[D min , D max ];

[0057] Step 3.3: Layer the three key parameters in their respective intervals, and set the number of stratification levels to s. Then the intervals of the three key parameters can be divided into s sub-intervals, which are:

[0058] [L min1 , L max1 ],[L min2 , L max2 ],…,[L mins , L maxs ];

[0059] [W min1 , W max1 ],[W min2 , W max2 ],…,[W mins , W maxs ];

[0060] [D min1 , D max1 ],[D min2 , D max2 ],…,[D mink , D maxk ];

[0061] Step 3.4, start the Monto-Carlo random number generation algorithm (i.e., Monte Carlo algorithm);

[0062] Step 3.5: Perform random sampling in each level interval, obtain s random sample points for each of the three parameters, and then use the random number seed generated by the Monte Carlo algorithm to randomly sample the sample points of L, W, and D to obtain a sample point set (i.e., the above-mentioned s groups of second sample points);

[0063] Step 3.6: Calculate the distance L between every two points in the sample point set ij , and obtain the minimum distance L accordingly min ;

[0064] Step 3.7: Determine the minimum judgment standard L of the sample space s (i.e. the above-mentioned preset minimum distance);

[0065] Step 3.8: Determine whether L is satisfied min <L s ; If yes, execute step 3.9; if no, output the sampling point set;

[0066] Step 3.9: Adjust the distance between the two points corresponding to the minimum distance, and return to execute step 3.6.

[0067] Through the above embodiment, after random sampling is performed to generate sampling results, the mode of minimizing the distance in the multidimensional vector space is improved so that the distribution of the s groups of first sample points finally obtained is more uniform and the subsequent fitting test effect is better.

[0068] Optionally, before determining the reconciliation parameter of the comprehensive evaluation index of the cold plate fin according to the obtained fitting test result, the method further includes: determining the target expression of the comprehensive evaluation index as: Among them, ES is the comprehensive evaluation index, Nu is the dimensionless temperature gradient, Δp is the flow resistance loss, Nua is the reference value of the dimensionless temperature gradient, Δpa is the reference value of the flow resistance loss, and α and β are both the harmonic parameters.

[0069] During the cooling process, the selection of cold plate parameters needs to consider both heat transfer effect and flow resistance loss, and the two factors are contradictory and mutually restraining. Generally speaking, the stronger the heat transfer effect, the greater the flow resistance, and the smaller the flow resistance, the worse the heat transfer effect. Therefore, it is necessary to find the optimal design direction and the best balance point during the cold plate design process, and then realize the cold plate parameters under specific heat dissipation requirements.

[0070] The descriptive parameter for microchannel cooling inside the cold plate is Nu, i.e., the dimensionless temperature gradient, and the descriptive parameter for the flow resistance of the coolant inside the cold plate is the pressure Δp. In order to accurately and quantitatively describe the comprehensive influence of Nu and pressure on the performance of the cold plate, this application selects the form of standardized parameter ratio as the evaluation index ES of the cold plate performance, and introduces exponential harmonic factors α and β for harmonization. The comprehensive performance of the cold plate is analyzed, and the comprehensive evaluation index ES is defined as: Where Nu and Δp are the dimensionless temperature gradient and flow resistance loss of the cold plate under a given design state, and Nua and Δpa are the dimensionless temperature gradient and flow resistance loss of the cold plate under a reference state. Generally, a specific design parameter is taken as the reference state.

[0071] Since different forms of cold plates have different index evaluation criteria in different application scenarios, the emphasis on hydraulic performance and thermal performance varies. In order to explore a flexible evaluation index and improve the universality of the index. The embodiment of the present application introduces exponential adjustment factors α and β in both the numerator and denominator of the comprehensive evaluation index. By adjusting the values ​​of α and β for cold plates of different forms to adapt to the index driving under different engineering application conditions, the accuracy of the ES index in predicting comprehensive performance is improved.

[0072] Optionally, performing a fitting test on the cold plate fin according to the first sample point of the s group, and determining the harmonic parameter of the comprehensive evaluation index of the cold plate fin according to the obtained fitting test result, includes: performing a fitting test on the cold plate fin according to the first sample point of the s group, and determining the heat exchange target temperature T of the cold plate fin i and cost parameter S i , wherein the heat exchange target temperature is used to indicate the heat exchange performance of the cold plate fin, the thermal performance includes the heat exchange performance, and the cost parameter is used to indicate the construction cost of the cold plate fin, T i is the heat exchange target temperature of the cold plate fin corresponding to the first sample point of the i-th group, S i is the cost parameter of the cold plate fin corresponding to the first sample point of the i-th group; determining the engineering priority sequence number C of the first sample point of the s-th group according to the engineering requirement information of the cold plate fin, the heat exchange target temperature and the cost parameter TSi , where C TSi is the engineering priority sequence number of the first sample point of the i-th group, i is a positive integer; the initial value α0 is set for the harmonic parameter α, and the initial value β0 is set for the harmonic parameter β to obtain the first harmonic parameter; a fitting test is performed based on the first sample point of the s group to calculate the dimensionless temperature gradient Nu of the cold plate fin i and flow resistance loss Δp i , wherein the fitting test results include: the heat exchange target temperature T i , the cost parameter Si , the dimensionless temperature gradient Nu i and the flow resistance loss Δp i , Nu i is the dimensionless temperature gradient of the cold plate fin corresponding to the first sample point of the i-th group, Δp i is the flow resistance loss of the cold plate fin corresponding to the first sample point of the i-th group; according to the first harmonic parameter, the target expression, the dimensionless temperature gradient Nu i and the flow resistance loss Δp i Calculate the first index value of the comprehensive evaluation index, and determine the index priority sequence number C of the s groups of first sample points according to the s first index values ESi , where C ESi is the index priority sequence number of the first sample point in the i-th group; according to C TSi With C ESi The first harmonic parameter is adjusted according to the absolute difference of , so as to obtain the harmonic parameter.

[0073] Optionally, the TSi With C ESi The method comprises: a second adjustment step: when the maximum value among the plurality of absolute differences is greater than the difference judgment standard, adjusting the first harmonic parameter according to the maximum value to obtain a second harmonic parameter; a calculation step: calculating the second harmonic parameter, the target expression, the dimensionless temperature gradient Nu according to the second harmonic parameter, the target expression, the dimensionless temperature gradient Nu, and the dimensionless temperature gradient Nu. i and the flow resistance loss Δp i Calculate the second index value of the comprehensive evaluation index, and re-determine the index priority sequence number C of the s groups of first sample points according to the s second index values ESi ; Determination step: recalculate multiple absolute differences, and redetermine the maximum value among the multiple absolute differences, and determine the size of the maximum value and the difference judgment standard; loop the second adjustment step, the calculation step and the determination step until the maximum value is less than or equal to the difference judgment standard, and determine the final second harmonic parameter as the harmonic parameter.

[0074] In the embodiment of the present application, by introducing the comprehensive evaluation index ES of the cold plate performance, the thermal performance and hydraulic performance of the cold plate are integrated and evaluated, taking into account the factors affecting the operation performance of the cold plate. For the reconciliation factors α and β in the mathematical description of the ES index, the engineering priority number and the index priority number of different cold plate parameter design states are compared through the priority sequence method, and α and β are iteratively solved according to the extreme difference, so as to determine the specific function form of ES.

[0075] The workflow of the priority sequence method is as follows: Figure 4 As shown, the following steps are included:

[0076] Step 4.1: First, perform (L i , W i , D i ) simulation calculation, numerical simulation of the corresponding cold plate design state, and obtain the corresponding heat exchange target temperature T i At the same time, the engineering cost of the cold plate state of each sample point is quantitatively converted to obtain the cost parameter S i ;

[0077] Step 4.2: Calculate the corresponding Nu based on the simulation results i and Δp i ;

[0078] Step 4.3: Comprehensive T i and S i The value of the project priority sequence number C of each sample point is determined in combination with the project requirements. TSi ;

[0079] Step 4.4, assign initial values ​​α0 and β0 to the harmonic parameters α and β;

[0080] Step 4.5: Calculate the initial value of the ES indicator E Sji according to the initial value of the harmonic parameter, and determine the corresponding indicator priority sequence number C according to the size of E Sji ESji , j = 0 (where j represents the number of iterations);

[0081] Step 4.6: Calculate C TSi and C ESi The absolute difference D TEi =|C TSi -C ESi |, and select the maximum absolute difference D from all sample points TEimax , compare the maximum value with the difference judgment standard ε. If the maximum value is greater than the difference judgment standard, adjust the values ​​of the harmonic parameters α and β according to the maximum value, update the j value, and then iterate steps 4.5-4.6; if the maximum value is less than the difference judgment standard, execute step 4.7;

[0082] Step 4.7: Confirm that the iteration is complete and output the final harmonic parameters.

[0083] Through the above embodiments, the dual-objective optimization from Nu to Δp is simplified to the dimension reduction of the single-objective optimization for ES. After the dimension reduction is completed, the single-objective optimization process can be carried out according to the ES index. The final optimized result is a single point rather than a point set, which eliminates the need for secondary optimization of the optimization point set after the dual-objective optimization is completed. It has strong engineering operability, and the obtained optimization point index combination fully considers the influence of heat dissipation and flow resistance in the specific engineering application background, and the state is more intuitive.

[0084] It should be noted that after determining the harmonic parameters, the complete mathematical description of ES can be obtained, thereby reducing the dimension of the dual-objective optimization problem for Nu and Δp to a single-objective optimization of ES parameters. By using optimization algorithms such as neural networks, response surface methods, and genetic networks, the single-objective solution of ES can be achieved. The resource requirements for the Pareto solution of single-objective optimization are much lower than those of dual-objective optimization. At the same time, the two index factors of thermal and hydraulic performance are fully considered, thereby improving the accuracy and efficiency of the optimization algorithm at the same time. While ensuring the accuracy of the solution, the demand for computing resources and the solution steps are simplified, with stronger engineering operability, so that cold plates with strong engineering adaptability can be designed more quickly and efficiently.

[0085] Through the description of the above implementation methods, those skilled in the art can clearly understand that the method according to the above embodiment can be implemented by means of software plus a necessary general hardware platform, and of course by hardware, but in many cases the former is a better implementation method. Based on this understanding, the technical solution of the present application, or the part that contributes to the prior art, can be embodied in the form of a software product, which is stored in a storage medium (such as ROM / RAM, magnetic disk, optical disk), and includes a number of instructions for a terminal device (which can be a mobile phone, computer, server, or network device, etc.) to execute the methods described in each embodiment of the present application.

[0086] In this embodiment, a device for determining a comprehensive evaluation index is also provided. Figure 6 is a structural block diagram of a device for determining a comprehensive evaluation index according to an embodiment of the present application, such as Figure 6 As shown, the device comprises:

[0087] The sampling module 62 is used to sample the fin parameters of the cold plate fins to obtain s groups of first sample points, wherein the fin parameters include: fin length, fin width and fin spacing, and the s groups of first sample points are used to indicate a numerical combination of multiple fin parameters, and s is a positive integer;

[0088] A testing module 64 is used to perform a fitting test on the cold plate fin according to the s group of first sample points, and determine a harmonic parameter of a comprehensive evaluation index of the cold plate fin according to the obtained fitting test result, wherein the comprehensive evaluation index is used to comprehensively evaluate the thermal performance and hydraulic performance of the cold plate fin, and the harmonic parameter is used to indicate the importance of the dimensionless temperature gradient and flow resistance loss of the cold plate fin, the dimensionless temperature gradient is used to evaluate the thermal performance of the cold plate fin, and the flow resistance loss is used to evaluate the hydraulic performance of the cold plate fin;

[0089] The determination module 66 is used to determine the comprehensive evaluation index of the cold plate fin according to the blending parameter.

[0090] Through the above device, the fin parameters of the cold plate fin are first sampled to obtain s groups of first sample points, the fin parameters include: fin length, fin width and fin spacing, the s groups of first sample points are used to indicate different combinations of fin parameters, s is a positive integer; according to the s groups of first sample points, the cold plate fin is fitted and tested, and the harmonic parameter of the comprehensive evaluation index for comprehensively evaluating the thermal performance and hydraulic performance of the cold plate fin is determined according to the fitting test result. The harmonic parameter is used to indicate the importance of the dimensionless temperature gradient and flow resistance loss of the cold plate fin, and the dimensionless temperature gradient is used to evaluate the thermal performance. The flow resistance loss is used to evaluate the hydraulic performance; finally, the comprehensive evaluation index of the cold plate fin is determined according to the harmonized parameter; using the above scheme, a dual-objective dimensionality reduction method is proposed, and a new comprehensive evaluation index is created according to the thermal performance index and the hydraulic performance index, so that the cold plate can be optimized accurately and quickly with a single objective through the comprehensive evaluation index to determine the optimal state point; thereby solving the problem in the related technology that the design optimization process of the liquid-cooled cold plate is difficult to determine the optimal state point by simultaneously performing dual-objective optimization design in the two directions of minimizing the flow resistance and maximizing the heat exchange performance.

[0091] Optionally, the sampling module 62 is further used to respectively determine the value ranges of the multiple fin parameters, and stratify the multiple value ranges into s layers, wherein the s sub-intervals obtained after stratification of each value range are equal in length, and the multiple fin parameters correspond one-to-one to the multiple value ranges; random sampling is performed in the s sub-intervals of each fin parameter to obtain s random sampling points corresponding to each fin parameter; a combination step: among the s random sampling points corresponding to each fin parameter, a first random sampling point is selected by a random number seed, and multiple first random sampling points are combined to obtain a group of sample points, wherein the multiple fin parameters correspond one-to-one to the multiple first random sampling points; the combination step is executed s times in a loop to obtain s groups of second sample points, wherein the s second random sampling points corresponding to the same fin parameter in the s groups of second sample points are different from each other; the s groups of first sample points are determined according to the s groups of second sample points.

[0092] Optionally, the above-mentioned sampling module 62 is also used to use multiple first random sampling points of the second sample point as spatial coordinates, calculate the spatial distances between any two groups of second sample points respectively, and obtain multiple first spatial distances; determine the size of the second spatial distance among the multiple first spatial distances and a preset minimum distance, wherein the second spatial distance is the smallest among the multiple first spatial distances; when the second spatial distance is greater than the preset minimum distance, determine the s groups of second sample points as the s groups of first sample points.

[0093] Furthermore, the above-mentioned sampling module 62 is also used for the first adjustment step: when the second spatial distance is less than the preset minimum distance, multiple first random sampling points of the two groups of second sample points corresponding to the second spatial distance are adjusted so that the spatial distance between the two groups of second sample points corresponding to the second spatial distance is greater than the preset minimum distance; update step: recalculate the spatial distance between any two groups of second sample points in the s groups of second sample points, and update the second spatial distance according to the calculation result; and loop the first adjustment step and the update step until the second spatial distance is greater than the preset minimum distance.

[0094] Optionally, the test module 64 is further used to determine the target expression of the comprehensive evaluation index as: Wherein, ES is the comprehensive evaluation index, Nu is the dimensionless temperature gradient, Δp is the flow resistance loss, Nu a is the reference value of the dimensionless temperature gradient, Δp a is the reference value of the flow resistance loss, and α and β are the harmonic parameters.

[0095] Optionally, the test module 64 is further used to perform a fitting test on the cold plate fin according to the s groups of first sample points to determine the heat exchange target temperature T of the cold plate fin. i and cost parameter S i , wherein the heat exchange target temperature is used to indicate the heat exchange performance of the cold plate fin, the thermal performance includes the heat exchange performance, and the cost parameter is used to indicate the construction cost of the cold plate fin, T i is the heat exchange target temperature of the cold plate fin corresponding to the first sample point of the i-th group, Si is the cost parameter of the cold plate fin corresponding to the first sample point of the i-th group; the engineering priority sequence number C of the first sample point of the s-th group is determined according to the engineering requirement information of the cold plate fin, the heat exchange target temperature and the cost parameter TSi , where C TSi is the engineering priority sequence number of the first sample point of the i-th group, i is a positive integer; the initial value α0 is set for the harmonic parameter α, and the initial value β0 is set for the harmonic parameter β to obtain the first harmonic parameter; a fitting test is performed based on the first sample point of the s group to calculate the dimensionless temperature gradient Nu of the cold plate fin i and flow resistance loss Δp i , wherein the fitting test results include: the heat exchange target temperature T i , the cost parameter S i , the dimensionless temperature gradient Nu i and the flow resistance loss Δp i , Nu i is the dimensionless temperature gradient of the cold plate fin corresponding to the first sample point of the i-th group, Δp i is the flow resistance loss of the cold plate fin corresponding to the first sample point of the i-th group; according to the first harmonic parameter, the target expression, the dimensionless temperature gradient Nu i and the flow resistance loss Δp i Calculate the first index value of the comprehensive evaluation index, and determine the index priority sequence number C of the s groups of first sample points according to the s first index values ESi , where C ESi is the index priority sequence number of the first sample point in the i-th group; according to C TSi With C Esi The first harmonic parameter is adjusted according to the absolute difference of , so as to obtain the harmonic parameter.

[0096] Optionally, the test module 64 is further configured to cyclically execute the second adjustment step, the calculation step, and the determination step until the maximum value is less than or equal to the difference criterion, and the second harmonic parameter finally obtained is determined as the harmonic parameter; wherein the second adjustment step includes: when the maximum value among the multiple absolute differences is greater than the difference criterion, adjusting the first harmonic parameter according to the maximum value to obtain the second harmonic parameter; the calculation step includes: according to the second harmonic parameter, the target expression, the dimensionless temperature gradient Nu i and the flow resistance loss Δp i Calculate the second index value of the comprehensive evaluation index, and re-determine the index priority sequence number C of the s groups of first sample points according to the s second index values ESi The determination step includes: recalculating multiple absolute difference values, and re-determining the maximum value among the multiple absolute difference values, and determining the size of the maximum value and the difference judgment standard.

[0097] It should be noted that the above modules can be implemented by software or hardware. For the latter, it can be implemented in the following ways, but not limited to: the above modules are all located in the same processor; or the above modules are located in different processors in any combination.

[0098] An embodiment of the present application further provides a computer-readable storage medium, in which a computer program is stored, wherein the computer program is configured to execute the steps of any of the above method embodiments when running.

[0099] In an exemplary embodiment, the computer-readable storage medium may include, but is not limited to, various media that can store computer programs, such as a USB flash drive, a read-only memory (ROM), a random access memory (RAM), a mobile hard disk, a magnetic disk or an optical disk.

[0100] An embodiment of the present application further provides an electronic device, including a memory and a processor, wherein a computer program is stored in the memory, and the processor is configured to run the computer program to execute the steps in any one of the above method embodiments.

[0101] In an exemplary embodiment, the electronic device may further include a transmission device and an input / output device, wherein the transmission device is connected to the processor, and the input / output device is connected to the processor.

[0102] An embodiment of the present application further provides a computer program product, comprising a non-volatile computer-readable storage medium, wherein the non-volatile computer-readable storage medium stores a computer program product, and when the computer program is executed by a processor, the steps of the method described in each embodiment of the present application are implemented.

[0103] For specific examples in this embodiment, reference may be made to the examples described in the above embodiments and exemplary implementation modes, and this embodiment will not be described in detail herein.

[0104] Obviously, those skilled in the art should understand that the above modules or steps of the present application can be implemented by a general computing device, they can be concentrated on a single computing device, or distributed on a network composed of multiple computing devices, they can be implemented by a program code executable by a computing device, so that they can be stored in a storage device and executed by the computing device, and in some cases, the steps shown or described can be executed in a different order from that herein, or they can be made into individual integrated circuit modules, or multiple modules or steps therein can be made into a single integrated circuit module for implementation. Thus, the present application is not limited to any specific combination of hardware and software.

[0105] The above description is only the preferred embodiment of the present application and is not intended to limit the present application. For those skilled in the art, the present application may have various modifications and variations. Any modification, equivalent replacement, improvement, etc. made within the principles of the present application shall be included in the protection scope of the present application.

Claims

1. A method for determining a comprehensive evaluation index, characterized in that: include: Sampling the fin parameters of the cold plate fins to obtain s groups of first sample points, wherein the fin parameters include: fin length, fin width and fin spacing, and the s groups of first sample points are all used to indicate a numerical combination of multiple fin parameters, and s is a positive integer; A fitting test is performed on the cold plate fin according to the first sample point of the s group, and a harmonic parameter of a comprehensive evaluation index of the cold plate fin is determined according to the obtained fitting test result, wherein the comprehensive evaluation index is used to comprehensively evaluate the thermal performance and hydraulic performance of the cold plate fin, and the harmonic parameter is used to indicate the importance of the dimensionless temperature gradient and flow resistance loss of the cold plate fin, the dimensionless temperature gradient is used to evaluate the thermal performance of the cold plate fin, and the flow resistance loss is used to evaluate the hydraulic performance of the cold plate fin; The comprehensive evaluation index of the cold plate fin is determined according to the blending parameter.

2. The method for determining the comprehensive evaluation index according to claim 1, characterized in that: The step of sampling the fin parameters of the cold plate fins to obtain s groups of first sample points includes: Determine the value ranges of the plurality of fin parameters respectively, and divide the plurality of value ranges into s layers, wherein the s sub-intervals obtained after each value range is divided into layers have the same length, and the plurality of fin parameters correspond to the plurality of value ranges one by one; Performing random sampling in s subintervals of each fin parameter to obtain s random sampling points corresponding to each fin parameter; Combining step: selecting a first random sampling point from the s random sampling points corresponding to each of the fin parameters by a random number seed, and combining a plurality of the first random sampling points to obtain a group of sample points, wherein the plurality of the fin parameters correspond one-to-one to the plurality of the first random sampling points; The combining step is executed s times in a loop to obtain s groups of second sample points, wherein the s second random sampling points corresponding to the same fin parameter in the s groups of second sample points are different from each other; The s groups of first sample points are determined according to the s groups of second sample points.

3. The method for determining the comprehensive evaluation index according to claim 2, characterized in that: The determining the s groups of first sample points according to the s groups of second sample points comprises: Taking the first plurality of random sampling points of the second sample point as spatial coordinates, respectively calculating the spatial distances between any two groups of the second sample points, to obtain a plurality of first spatial distances; Determine a magnitude between a second spatial distance among the plurality of first spatial distances and a preset minimum distance, wherein the second spatial distance is the smallest among the plurality of first spatial distances; When the second spatial distance is greater than the preset minimum distance, the s groups of second sample points are determined as the s groups of first sample points.

4. The method for determining the comprehensive evaluation index according to claim 3, characterized in that: After determining the difference between the second spatial distance among the plurality of first spatial distances and the preset minimum distance, the method further includes: A first adjustment step: when the second spatial distance is less than the preset minimum distance, adjusting a plurality of the first random sampling points of the two groups of second sample points corresponding to the second spatial distance so that the spatial distance between the two groups of second sample points corresponding to the second spatial distance is greater than the preset minimum distance; Updating step: recalculating the spatial distance between any two groups of second sample points in the s groups of second sample points, and updating the second spatial distance according to the calculation result; The first adjusting step and the updating step are executed in a loop until the second spatial distance is greater than the preset minimum distance.

5. The method for determining the comprehensive evaluation index according to claim 1, characterized in that: Before determining the reconciliation parameter of the comprehensive evaluation index of the cold plate fin according to the obtained fitting test results, the method further includes: The target expression for determining the comprehensive evaluation index is: Wherein, ES is the comprehensive evaluation index, Nu is the dimensionless temperature gradient, Δp is the flow resistance loss, Nu a is the reference value of the dimensionless temperature gradient, Δp a is the reference value of the flow resistance loss, and α and β are the harmonic parameters.

6. The method for determining the comprehensive evaluation index according to claim 5, characterized in that: The step of performing a fitting test on the cold plate fin according to the s groups of first sample points, and determining the harmonic parameter of the comprehensive evaluation index of the cold plate fin according to the obtained fitting test result, includes: The cold plate fin is fitted with the first sample point of the s group to determine the heat exchange target temperature T of the cold plate fin. i and cost parameter S i , wherein the heat exchange target temperature is used to indicate the heat exchange performance of the cold plate fin, the thermal performance includes the heat exchange performance, and the cost parameter is used to indicate the construction cost of the cold plate fin, T i is the heat exchange target temperature of the cold plate fin corresponding to the first sample point of the i-th group, S i is the cost parameter of the cold plate fin corresponding to the first sample point of the i-th group; Determine the engineering priority sequence number C of the first sample point of the group s according to the engineering requirement information of the cold plate fin, the heat exchange target temperature and the cost parameter TSi , where C TSi is the engineering priority sequence number of the first sample point in the i-th group, where i is a positive integer; Setting an initial value α0 for the harmonic parameter α and setting an initial value β0 for the harmonic parameter β to obtain a first harmonic parameter; According to the first sample point of the s group, a fitting test is performed to calculate the dimensionless temperature gradient Nu of the cold plate fin. i and flow resistance loss Δp i , wherein the fitting test results include: the heat exchange target temperature T i , the cost parameter S i , the dimensionless temperature gradient Nui and the flow resistance loss Δp i , Nu i is the dimensionless temperature gradient of the cold plate fin corresponding to the first sample point of the i-th group, and Δpi is the flow resistance loss of the cold plate fin corresponding to the first sample point of the i-th group; According to the first harmonic parameter, the target expression, the dimensionless temperature gradient Nu i and the flow resistance loss Δpi to calculate the first index value of the comprehensive evaluation index, and determine the index priority sequence number C of the s groups of first sample points according to the s first index values ESi , where C ESi is the index priority sequence number of the first sample point in the i-th group; According to C Tsi With C Esi The first harmonic parameter is adjusted according to the absolute difference of , so as to obtain the harmonic parameter.

7. The method for determining the comprehensive evaluation index according to claim 6, characterized in that: According to C Tsi With C Esi The first harmonic parameter is adjusted according to the absolute difference of , to obtain the harmonic parameter, including: A second adjustment step: when the maximum value among the plurality of absolute differences is greater than the difference judgment standard, adjusting the first harmonic parameter according to the maximum value to obtain a second harmonic parameter; Calculation step: According to the second harmonic parameter, the target expression, the dimensionless temperature gradient Nu i and the flow resistance loss Δpi to calculate the second index value of the comprehensive evaluation index, and re-determine the index priority sequence number C of the s groups of first sample points according to the s second index values ESi ; Determining step: recalculating the plurality of absolute differences, and re-determining the maximum value among the plurality of absolute differences, and determining the magnitude of the maximum value and the difference judgment standard; The second adjustment step, the calculation step and the determination step are executed in a loop until the maximum value is less than or equal to the difference criterion, and the second harmonization parameter finally obtained is determined as the harmonization parameter.

8. A device for determining a comprehensive evaluation index, characterized in that: include: A sampling module, used for sampling the fin parameters of the cold plate fins to obtain s groups of first sample points, wherein the fin parameters include: fin length, fin width and fin spacing, and the s groups of first sample points are used to indicate a numerical combination of multiple fin parameters, and s is a positive integer; A testing module, used for performing a fitting test on the cold plate fin according to the first sample point of the s group, and determining a harmonic parameter of a comprehensive evaluation index of the cold plate fin according to the obtained fitting test result, wherein the comprehensive evaluation index is used for comprehensively evaluating the thermal performance and hydraulic performance of the cold plate fin, and the harmonic parameter is used for indicating the importance of the dimensionless temperature gradient and flow resistance loss of the cold plate fin, the dimensionless temperature gradient is used for evaluating the thermal performance of the cold plate fin, and the flow resistance loss is used for evaluating the hydraulic performance of the cold plate fin; A determination module is used to determine the comprehensive evaluation index of the cold plate fin according to the blending parameter.

9. A computer-readable storage medium, characterized in that: The computer-readable storage medium stores a computer program, wherein the computer program implements the steps of the method according to any one of claims 1 to 7 when executed by a processor.

10. An electronic device comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, characterized in that: When the processor executes the computer program, the steps of the method according to any one of claims 1 to 7 are implemented.

Images